LINGUIST List 7.660

Sun May 5 1996

Disc: Ungrammaticality

Editor for this issue: Ann Dizdar <dizdartam2000.tamu.edu>

Directory

1. Karl Teeter, In self-defense, bis (or ter) ungrammaticality
2. sstraighbingsuns.cc.binghamton.edu ("H Stephen Straight (Binghamton University , SUNY)"), Disc: Ungrammaticality and the Myth of G

Dear Benji: Thanks a lot for your message and questions. I do try to
keep track of my mistakes and lapses, commoner since I turned 67, such as
failing to recognize early in this very same discussion that "the happy
scissors" is perfectly grammatical. But I don't mind my
mistakes and lapses as much as I mind being obscure, and that seems to be
my latest sin here.

	 Most actual utterances are ungrammatical. Most utterances, of
course, are grunts, groans, sighs, and the like. Furthermore, you are
right in emphasizing Bloomfield's point that no two utterances are ever the
same in the real world. This, of course, makes the enterprise of grammar
impossible, unless we are willing to become structuralists, which
Bloomfield makes possible by asserting that we must assume that "some
utterances are the same" (this is an assumption, since we know it to be
false). To Bloomfield, good physicalist that he was, no two
utterances are in fact the same, but we know we are aiming at grammar, and
hence this becomes the fundamental assumption of linguistics, the
philosophical assumption that linguistics is possible even if impossible
as a physical science.
		Sorry, Benji and friends, perhaps it was Bloomfield that
was being obscure and not me, but when Bloomfield got portentous enough
to call something the "fundamental assumption of linguistics" he was being
deeply serious and we'd better believe it! His "fundamental assumption"
is in fact the bedrock for structural linguistics and structuralism in
general. I could go on and on, but won't unless asked. Yours, kvt

	

On Tue, 30 Apr 1996, The Linguist List wrote:
> Date: Sun, 28 Apr 1996 22:22:00 PDT
> From: IBENAWJMVS.OAC.UCLA.EDU (benji wald )
> ... In any case, we have a
> paradox in that in some contexts the term "grammar" refers to something
> "real" and presumably natural and innate in some ways, and in other
> contexts, theoretical constructs about language data and the nature of
> language that are -- uhm, theoretical constructs of language data and...
> In practice it is not always easy to keep these two contexts of discussion
> apart.

As usual, Benji Wald's contribution forces us to examine our
assumptions. One alternative to the above that virtually all
linguists since Saussure have neglected is that "theoretical
constructs of language data" might be able to provide -- directly --
an account of something that is going in language users when they
perceive or produce language. Instead linguists have provided only
accounts that deliberately postpone that effort by constructing
"grammars" that provide (competing) structural descriptions of
language forms abstracted away from the processes whereby they might
be recognized or constructed. Saussure came close to rejecting this
process-neutral approach to language when he localized the proper
object of linguistic study as "la portion de'termine'e du circuit [de
parole] o`u une image auditive vient s'associer `a un concept" (CLG
1916/1976:31), and insisted that "le co^te' exe'cutif [parole] reste
hors de cause" (30).

Of course, Saussure's naive associationism cannot be resuscitated and
- besides -- even Saussure failed to maintain an exclusive focus on
the receptive aspect of language and the "faculte' d'association et de
coordination" (29) that enables a listener to make sense of language
input in context. But that doesn't mean that such a focus cannot be
maintained. Nor does it mean that this (left/posterior) focus can't
(now that we know so much more about the anterior/posterior and
left/right cerebral dualities of language functions) be paired with
complementary but separate foci on the (right/posterior) processes
whereby a listener builds a global discursive representation on the
basis of item-by-item interpretations and the corresponding
(right/anterior and posterior) processes whereby a speaker organizes
complex linguistic intentions and arrives at specific means for
expressing them.

Finally, to the above four foci we could add an account of how the
simultaneous activation of all of these receptive and expressive
processes brings forth such fascinating things as anacoloutha,
ellipses, paraphrases, self-repairs, grammaticality judgments, and
other complex products of receptive/expressive interaction. On this
account, it is ironic that judgments of grammaticality -- which are
here identified as arguably the most complex, variable, and
problematic by-products of language processing -- have for so long
held center stage as the supposedly simplest, fixed, and unproblematic
building blocks for linguistic science.

Don't get me wrong. Grammars are great. Despite their problematic
epistemological and ontological status, they provide diverse and
invaluable insights into the logical features of the language system
as a whole, just as mathematical physics provides deep insights into
the nature of the physical universe. But localizing grammars in the
brain/mind makes no more sense than localizing Einstein's equations in
the ether. It's this reification of grammar that we may one day
reject as "the Myth of G". (See Straight, forthcoming, "Conduction
aphasia, specific language impairment, and the lessons of rare-event
linguistics". :-)

		Best.	 'Bye.	 Steve
 H Stephen Straight, Dir, Lgs Across the Curric, Binghamton U (SUNY)
 NFLC Mellon Fellow, Jan-Jun 96 VOX: 202/667-8100, FAX: 202/667-6907
 National Foreign Lang Ctr, 1619 Mass Ave NW, Washington, DC 20036
 <sstraighbingsuns.cc.binghamton.edu> ["sstraigh", not "sstraight"]